Author: [Bruce Zhou]
Affiliation: [Jota Machinery Composites Material Prepreg Solution]
Corresponding Author: [jotamachinery@gmail.com]
Published : December 08 , 2025

Abstract

Carbon–carbon (C/C) composites combine carbon fibres with a carbon matrix to deliver a rare combination of high-temperature stability, thermal shock resistance and friction performance. While their potential has been recognized for decades, widespread use remains constrained by high production costs and vulnerability to oxidation. This paper revisits a landmark 1993 Defence Science Journal article by Devi and Rama Rao, together with related carbon–carbon literature, and reframes it from a modern engineering perspective. The focus is on multidirectional preform architectures (3D–6D), liquid-phase densification with high-pressure carbonisation, oxidation protection strategies and full-scale aircraft brake disc development at the Defence Research and Development Laboratory (DRDL) in India.

We first outline the unique thermo-structural and tribological behaviour of carbon–carbon composites and place DRDL’s work in the international context of the early 1990s. We then describe the preform design logic, impregnation and densification routes, and graphitisation practices used to achieve densities around 1.7–1.8 g/cm³ with 3D/4D architectures. Measured material properties, thermal behaviour and brake dynamometer results (coefficient of friction 0.22–0.33, wear ≈8 µm/face/stop) are summarised and interpreted. Particular attention is given to the microstructure of resin-based, pitch-based and hybrid matrices, and to the role of oxidation damage in limiting service life.

The discussion translates DRDL’s experience into design rules that remain relevant for current carbon–carbon and C/SiC developers: when multidirectional preforms are worth the manufacturing complexity, how high-pressure carbonisation can cut densification cycles, and why friction film formation governs brake reliability more than any single coupon property. Remaining gaps – detailed oxidation kinetics, life prediction and industrial cost models – are highlighted as opportunities for present-day research and process innovation.

Keywords

carbon–carbon composites; multidirectional preforms; aircraft brake discs; high-pressure carbonisation; oxidation protection; PAN-based carbon fibre; pitch-based matrices; C/SiC hybrids; tribology; aerospace thermal-structural materials

1. Introduction

Carbon–carbon composites occupy a narrow but crucial corner of structural materials technology. By embedding carbon fibres in a carbon matrix, engineers obtain components that do not melt, tolerate steep temperature gradients, and keep their strength where metals, polymers and many ceramics have already failed. This combination makes C/C the material of choice for re-entry nose tips, rocket nozzles, aero-engine hardware and high-energy aircraft brake discs.

The same advantages, however, come with serious penalties. Above about 500 °C in oxygen-bearing atmospheres, carbon begins to oxidise. Densifying porous fibre preforms into sound, high-density composites requires multiple impregnations and heat treatments. Costs rise quickly, and the technology typically remains in the hands of a small group of defence and aerospace organisations.

By the early 1990s, the United States, Europe and the former Soviet Union had already fielded C/C brake systems and thermal-structural components. At that time, the Defence Research and Development Laboratory (DRDL) in India began building its own carbon–carbon capability, targeting multidirectional structures and aircraft brakes. Their 1993 paper “Carbon–Carbon Composites – An Overview” documents, in compact form, how a national laboratory progressed from 3D preforms to full-scale dynamometer testing of C/C brake discs.

This paper revisits that work as practising engineers. We are less interested in repeating the historical narrative and more focused on extracting the practical rules hidden in the original data: how multidirectional architectures were designed, how high-pressure carbonisation changed the economics of thick parts, how matrix choice influenced microstructure and friction, and how brake discs were qualified. We also link those lessons to the broader carbon–carbon composite literature and to today’s renewed interest in C/C and C/SiC for hypersonic and fifth-generation aerospace systems.

2. Literature Review

2.1 Global development of carbon–carbon composites

By the time DRDL published its overview, carbon–carbon composites had already undergone several decades of development. Early work in the United States and Europe centred on rocket vanes and re-entry hardware, with later expansion into aircraft brakes and industrial furnace components. Classic texts and reviews from that period emphasise several constant themes:

• Thermal and mechanical behaviour
  C/C retains integrity to temperatures approaching 3000 °C in inert atmospheres, with low density and low thermal expansion. Failure is governed by fibre pull-out, debonding and crack deflection, giving higher toughness than monolithic ceramics in similar environments.
• Processing routes
  Two main matrix densification routes dominate: chemical vapour infiltration/deposition (CVI/CVD) and repeated liquid impregnation with subsequent pyrolysis. CVI/CVD yields high-purity pyrolytic carbon matrices, particularly suitable for thin structures and brake discs. Liquid routes using phenolic resins, pitch or combinations of both are better suited to thicker, bulkier parts but demand multiple cycles.
• Oxidation as the limiting factor
  Across sources, oxidation consistently appears as the primary lifetime constraint. The consensus solution combines matrix design, microstructure control and surface or in-depth oxidation protection systems, including SiC coatings, glassy sealants and inhibitor-filled matrices.
• Cost and market constraints
  Despite outstanding performance, C/C remains confined to niche applications. Densification cycles, high graphitisation temperatures and complex coatings keep component costs high and restrict commercial adoption.

These international observations provide the backdrop for DRDL’s efforts.

2.2 DRDL’s contribution: multidirectional preforms and brake discs

The 1993 Defence Science Journal paper from DRDL stands out because it brings all elements of the carbon–carbon chain under one roof: design and manufacture of multidirectional preforms, liquid-phase densification with high-pressure carbonisation, oxidation protection trials and full-scale brake dynamometer testing. The authors do not simply describe individual process steps; they show how each decision — fibre type, preform architecture, resin or pitch selection, heating rate — affects final performance.

Key differentiating aspects include:

• The move from 2D laminates to 3D, 4D and higher-order preforms, aimed at eliminating delamination and increasing through-thickness strength and thermal conductivity.
• The use of high-pressure carbonisation (≈1000 bar) to raise pitch carbon yield to about 90 % and drastically reduce the number of impregnation–pyrolysis cycles needed for thick components.
• A side-by-side comparison of resin-based, pitch-based and hybrid matrices, linking microstructure directly to fracture features, oxidation behaviour and wear.
• Systematic aircraft brake development, including inertia dynamometer testing of C/C rotor–stator stacks under realistic braking energies, with careful monitoring of friction coefficient and wear.

For practitioners interested in translating laboratory C/C technology into certified hardware, this integrated approach is at least as important as any single material property.

3. Methodology

This paper is a critical literature review with an engineering focus rather than a conventional experimental study. The methodology combines three strands:

1. Primary technical source
   The main evidence base is the Defence Science Journal article by Devi and Rama Rao (1993), which we treat as a detailed case study of carbon–carbon composite development at DRDL. All process windows, composition ranges, property values and brake test data are drawn from this source.
2. Contextual carbon–carbon literature
   Selected monographs and reviews on carbon–carbon composites are used to cross-check and place DRDL’s results in a wider context. Wherever the DRDL data align with or deviate from global experience, we comment explicitly.
3. Engineering interpretation
   The data are reorganised according to process–structure–property–performance logic familiar to composite engineers:
   • Preform architecture and fibre selection
   • Matrix impregnation and densification (including high-pressure carbonisation)
   • Heat treatment and graphitisation
   • Oxidation protection concepts
   • Coupon properties and full-scale brake performance

Rather than reproducing every table, we focus on trends, typical ranges and design-relevant contrasts. The goal is to translate the original work into practical guidance: where multidirectional preforms justify their cost, when to use pitch versus resin, and how to read dynamometer curves in the context of C/C tribology.

4. Results

4.1 Multidirectional preforms: 3D and 4D architectures

DRDL’s first step beyond conventional laminates was the development of multidirectional preforms. The motivation was clear: aircraft brakes and thermal-structural components see complex stress states and steep thermal gradients; 2D lay-ups are vulnerable to delamination and poor through-thickness conductivity.

The authors describe:

• 3D orthogonal preforms with straight bundles in the x, y and z directions.
• 4D and higher-order preforms where additional diagonal yarns are introduced to stiffen and strengthen inclined planes and to further suppress delamination.

Design variables include the number of bundles per location, spacing, local fibre volume fraction and target bulk density. DRDL reports successful manufacture of 3D–6D blocks and cylinders using manual and semi-automated looms. These preforms provide the skeleton for both bulk structural parts and brake components.

Measured characteristics for representative multidirectional C/C include:

• 3D PAN-based C/C
  • Total fibre volume fraction ~44 %, with Z-direction fraction around 27–28 %.
  • Density ≈1.8 g/cm³.
  • Porosity ~7–12 %.
  • Tensile moduli from roughly 75 to 96 GPa depending on direction.
  • Flexural and compressive strengths in the 80–100 MPa range.
• 4D rayon-based C/C
  • Total fibre volume fraction ~38 %, with lower Z-direction content (~14 %).
  • Similar densities but somewhat different anisotropy and toughness.

Although these numbers are modest compared with later aerospace-grade C/C, they represent a balanced compromise between isotropy, manufacturability and cost for the intended applications.

4.2 Densification and high-pressure carbonisation

The DRDL team investigated both gas-phase and liquid-phase densification routes but ultimately focused on liquid impregnation with high-pressure carbonisation for thick parts.

Key observations include:

• Atmospheric pressure processing
  • When using pitch or resin at atmospheric pressure, achieving densities around 1.7 g/cm³ required approximately 15–17 impregnation–carbonisation cycles.
  • Gas evolution during pyrolysis led to noticeable microcracking and open porosity, especially in regions where the matrix was tightly constrained by strong fibre–matrix adhesion.
• High-pressure carbonisation (≈1000 bar)
  • Under high pressure, pitch carbon yield increased from about 50 % to roughly 90 %, as gas evolution was suppressed and the matrix was forced into remaining porosity.
  • With an optimised sequence of atmospheric and high-pressure steps, target densities could be reached in about three cycles.
  • Microstructural examination showed denser matrices and fewer large pores, with variations depending on the use of resin skeletons, pitch impregnation or hybrids.

DRDL concluded that, for thick C/C components, high-pressure carbonisation is one of the few practical ways to reduce production time and cost while maintaining acceptable microstructure.

4.3 Matrix microstructures: resin, pitch and hybrids

The paper offers a clear comparison of three matrix concepts:

1. Resin-derived matrix (R)
   • Phenolic or other high-char resins create an initially isotropic carbon skeleton with relatively fine porosity.
   • During carbonisation, shrinkage leads to microcracks, particularly in matrix regions constrained by fibres.
   • Strong fibre–matrix bonding can drag fibres and cause damage; very weak bonding leaves interfacial gaps.
2. Pitch-derived matrix (P)
   • Pitch produces a more lamellar, graphitic microstructure, especially in bulk regions and between filaments.
   • Densification is faster, and thermal conductivity can be significantly higher after sufficient graphitisation.
   • However, highly graphitic matrices tend to be softer and can show higher wear under frictional loading.
3. Hybrid skeleton + pitch (R+P)
   • A resin-derived skeleton is first formed, then further densified with pitch.
   • Bulk matrix regions show lamellar structure, while the matrix near fibres remains less graphitised.
   • The result is a compromise: somewhat higher residual porosity than full-pitch systems but improved crack control and more balanced mechanical behaviour.

Micrographs in the original work clearly show these differences and link them to fracture surfaces and oxidation paths. For brake discs, the balance between hardness, toughness and thermal conductivity is particularly important.

4.4 Graphitisation and thermal properties

Graphitisation at elevated temperatures completes the densification process and sets the thermal properties. DRDL found that:

• Too rapid heating during graphitisation, especially through critical temperature ranges, caused delamination and internal damage.
• Adopting very low heating rates (on the order of a few degrees per hour) and carefully controlling the final treatment temperature allowed the team to avoid structural damage and reach the desired microstructure.
• Increasing the graphitisation temperature, particularly for pitch-rich matrices, led to substantial increases in thermal conductivity and changes in specific heat. Raising treatment temperature from about 2000 °C to 2600 °C roughly doubled or tripled thermal conductivity in some cases, especially along fibre directions.

These trends are consistent with the growth and alignment of graphitic domains under higher thermal treatment, and they are directly relevant for components that must manage both mechanical load and heat flux.

4.5 Brake disc tribology and dynamometer performance

The culmination of DRDL’s work was the development and testing of C/C aircraft brake discs. Using an inertia dynamometer, the team evaluated both “rosette” and plain lay-up designs under realistic braking energies.

Key results include:

• First-generation rosette lay-up
  • Coefficient of friction μ ≈ 0.20–0.26 under normal and overload conditions.
  • Wear rate around 8 µm per friction face per stop.
  • Stable behaviour but room for improvement in friction level and thermal management.
• Plain lay-up
  • Lower friction (μ ≈ 0.13–0.18), considered insufficient for the target aircraft application.
  • Higher wear and less stable tribological response.
• Improved batch after process optimisation
  • With higher graphitisation, improved thermal conductivity and adjusted fibre volume fraction, the next generation of discs achieved μ ≈ 0.22–0.33 over a series of 30 normal stops.
  • Stop times and friction remained stable, indicating the formation of a more consistent friction film and better heat distribution.

Microscopic examination of used discs showed oxidation of the matrix near the friction surface, with fibres protruding, and confirmed that oxidation depth over multiple stops remained within manageable limits when suitable surface protection and paints were used.

5. Discussion

5.1 Value and cost of multidirectional preforms

From an engineering standpoint, multidirectional preforms are justified whenever:

• The structure is exposed to complex, three-dimensional stress states.
• Through-thickness thermal conductivity and mechanical strength are critical.
• Delamination or ply-by-ply spalling would be unacceptable failure modes.

The DRDL experience confirms that 3D and 4D architectures can deliver more isotropic properties and greater damage tolerance than 2D laminates. However, the added complexity in weaving, preform inspection and impregnation means they should be reserved for genuinely demanding applications: brake rotors, nozzles and load-bearing thermal shields, rather than all C/C components.

5.2 High-pressure carbonisation as a cost lever

The comparison between atmospheric and high-pressure carbonisation is particularly instructive. Cutting the number of densification cycles from roughly seventeen to three is not a marginal improvement; it fundamentally changes the economics of thick carbon–carbon structures.

For organisations planning to scale up C/C production, DRDL’s findings suggest that investment in high-pressure facilities should be evaluated not only on equipment cost but on:

• Reduced labour and furnace time per component.
• Lower scrap rates due to more uniform densification.
• Potential for tailored microstructures by adjusting pressure, temperature and impregnant composition.

The data also highlight that pressure must be managed together with resin or pitch chemistry; not every matrix precursor will respond in the same favourable way.

5.3 Matrix selection and friction performance

The side-by-side evaluation of resin-based, pitch-based and hybrid matrices under brake loading brings out a core principle: tribological performance cannot be predicted from tensile or flexural coupons alone.

Pitch-rich systems, especially when highly graphitised, provide:

• Higher thermal conductivity, which helps to spread and dissipate braking heat.
• A more graphitic friction layer, influencing μ and wear.

Resin-rich systems, in contrast, tend to be:

• Harder and often tougher in bulk, but
• More prone to microcracking during carbonisation and possibly less favourable for friction film stability.

Hybrid approaches, such as resin skeletons followed by pitch densification, give designers another degree of freedom: they can tune surface behaviour and subsurface toughness independently, then validate the combination on the dynamometer.

5.4 Oxidation and lifetime

Both the original paper and later work underline that oxidation remains a primary constraint on C/C in oxidative environments. DRDL’s micrographs of used brake discs show matrix recession and exposed fibres near the surface, with oxidation depth of a few millimetres after multiple stops.

From a design perspective, this means:

• Surface coatings, paints and C/SiC hybridisation are not optional extras but essential elements of the material system.
• Component geometry and ventilation must be designed to limit local oxygen supply and manage temperature peaks.
• Inspection and maintenance intervals should be defined based on oxidation depth rather than simple operating hours.

Long-term oxidation kinetics, especially under variable humidity and temperature cycling, were not fully characterised in the early work and remain a fertile area for modern testing and modelling.

5.5 Implications for current C/C and C/SiC development

Although the original DRDL study predates nano-reinforced matrices, modern CVI modelling and advanced ceramic fibres, its practical lessons remain relevant:

• Architecture first: choose 2D, 3D or higher-order preforms based on actual load paths and thermal gradients, not on historical habit.
• Densification strategy: consider high-pressure carbonisation or other process intensification methods whenever component thickness and required density would otherwise demand many cycles.
• Matrix engineering: treat resin, pitch and hybrids as separate design tools, matching them to the friction and thermal environment of the application.
• System perspective: think in terms of “C/C plus protection plus environment”, especially for aerospace brakes and hot structures, rather than viewing the composite in isolation.

These principles apply equally to new C/C, C/SiC and related hybrid materials being developed today for hypersonic vehicles, launch systems and high-performance braking.

6. Conclusion

Re-examining DRDL’s early work on carbon–carbon composites shows how a focused national programme can build a complete technology chain, from multidirectional preforms to qualified aircraft brake discs. The technical achievements are clear:

• Design and manufacture of 3D–6D carbon preforms with tailored fibre volume fractions and near-isotropic behaviour where needed.
• Implementation of liquid-phase densification strategies using resin, pitch and hybrids, combined with high-pressure carbonisation to raise pitch yield and reduce cycle count.
• Controlled graphitisation that balances thermal conductivity and mechanical integrity.
• Development and dynamometer validation of C/C brake discs with friction coefficients in the 0.22–0.33 range and acceptable wear under realistic energy loads.

Equally important are the underlying design rules that still hold today: architecture must match the load path, densification strategy governs cost, matrix microstructure controls friction, and oxidation protection defines lifetime. For engineers working on modern carbon–carbon or C/SiC systems, these lessons provide a grounded starting point that can be combined with contemporary tools such as multiscale modelling, advanced coatings and nano-modified matrices.

Future work should deepen the understanding of oxidation kinetics, build reliable life prediction models for C/C in complex service environments, and refine cost models that reflect both process intensification and supply-chain constraints. By connecting historical case studies such as DRDL’s with current research, the community can move carbon–carbon composites beyond niche status and into a broader range of high-temperature, high-value applications.

References

1. Devi, G. R.; Rama Rao, K. Carbon–Carbon Composites – An Overview. Defence Science Journal, 1993, 43(4), 369–383.
2. Savage, G. Carbon–Carbon Composites. Chapman & Hall, 1993.
3. Windhorst, T.; Blount, G. Carbon–Carbon Composites: A Summary of Recent Developments and Applications. Materials & Design, 1997, 18(1), 11–15.
4. Additional contemporary reviews and handbooks on carbon–carbon and C/SiC composites, tribology of aircraft brakes, and oxidation protection systems, consulted to cross-check trends and parameter ranges cited in this article.